What are the hardware requirements for QED Nano?

QED Nano requires approximately 8.1 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed QED Nano and how large is it?

QED Nano is developed by lm-provers, a model with 4b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate QED Nano?

QED Nano is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

QED Nano by lm-provers — VRAM 8.1GB, 256K context

Name: QED Nano
Author: lm-provers

QED Nano is an open-source language model by lm-provers. Features: 4b LLM, VRAM: 8.1GB, Context: 256K, License: apache-2.0, LLM Explorer Score: 0.27.

Arxiv:2602.03773 Base model:finetune:lm-provers... Base model:lm-provers/qed-nano... Conversational Dataset:lm-provers/fineproofs-... Deploy:azure En Endpoints compatible Eval-results Qwen3 Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/lm-provers/QED-Nano

QED Nano Parameters and Internals

LLM Name	QED Nano
Repository 🤗	https://huggingface.co/lm-provers/QED-Nano
Base Model(s)	lm-provers/QED-Nano-SFT lm-provers/QED-Nano-SFT
Model Size	4b
Required VRAM	8.1 GB
Updated	2026-05-16
Maintainer	lm-provers
Model Type	qwen3
Model Files	5.0 GB: 1-of-2 3.1 GB: 2-of-2
Supported Languages	en
Model Architecture	Qwen3ForCausalLM
License	apache-2.0
Context Length	262144
Model Max Length	262144
Transformers Version	4.51.1
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151936
Torch Data Type	bfloat16
Errors	replace

Rank the QED Nano capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to QED Nano

Best Alternatives	Context / RAM	Downloads	Likes
FastContext 1.0 4B SFT	256K / 8.1 GB	5735	357
GRPO 4 70	256K / 8.1 GB	11	0
Qwen3 4B Instruct 2507	256K / 8.1 GB	3109972	911
Fable Traces	256K / 8.1 GB	316	210
Lightning 4B	256K / 8.1 GB	13	6
...fetensors.chatgpt GPT Codex V2	256K / 8.1 GB	666	2
FastContext 1.0 4B RL	256K / 8.1 GB	4559	61
Qwen3 4B Thinking 2507	256K / 8.1 GB	291023	606
...rs.chatgpt GPT Gpt5.1 Thinking	256K / 8.1 GB	435	3
Qwen3 4B Instruct 2507 FP8	256K / 5.2 GB	1100744	78

Note: green Score (e.g. "73.2") means that the model is better than lm-provers/QED-Nano.